Random hangs and freezes Mojave

[iCezar 1]

UPDATE: adding here a link to my EFI and "RunMe.app" output 

https://1drv.ms/f/s!AijwqQbUnoN1gsZK096vFcTQunbz1A

 

Hi everyone.
Ever since installing Mojave I've been experiencing extreme slowness. No idea what's causing it. Everything worked fine under High Sierra.
I am attaching troubleshooting files and details here.
Any support is appreciated.

Geekbench score in High Sierra. 

https://browser.geekbench.com/v4/cpu/9880925 

I have some other issues such as sleep not working but for now I'd at least like it to have the same speed as it did in High Sierra.

My configuration:

• AsRock x79 Extreme 4
• Xeon E5-1650 v2 - OC to 4.3GHz
• Kingston 16 GB RAM DDR3 1600 ECC - OC to 1866 MHz
• Asus GTX 760 mini 2 GB
• USB 3.0 add-in card FL1100 (Inateck KT4006)
• Samsung EVO 850 256GB SSD

 

kextstat|grep -y acpiplat
 14    1 0xffffff7f82912000 0x9c000    0x9c000    com.apple.driver.AppleACPIPlatform (6.1) 4423FFAE-7525-3F74-9E9D-E87EF1D43627 <13 12 11 7 6 5 4 3 1>

 

kextstat|grep -y appleintelcpu
   35    0 0xffffff7f83180000 0x2a000    0x2a000    com.apple.driver.AppleIntelCPUPowerManagement (220.0.0) B8FECE19-759D-37D3-9615-E3C6C7F5F222 <7 6 5 4 3 1>
   47    0 0xffffff7f8320e000 0x3000     0x3000     com.apple.driver.AppleIntelCPUPowerManagementClient (220.0.0) 14247E5A-E9E1-338E-B86E-5525A4520891 <7 6 5 4 3 1>

 

kextstat|grep -y applelpc
  123    0 0xffffff7f829d3000 0x3000     0x3000     com.apple.driver.AppleLPC (3.1) 2D0491D1-D130-346C-BCA3-BBC8A745FD00 <111 12 5 4 3>

 

kextstat|grep -y applehda
  119    1 0xffffff7f82524000 0x1d000    0x1d000    com.apple.driver.AppleHDAController (282.10) 1E3D72E3-0918-39C4-B0E7-4561D7915236 <118 108 88 12 7 6 5 4 3 1>
  156    0 0xffffff7f826a7000 0xb7000    0xb7000    com.apple.driver.AppleHDA (282.10) 88DA0A6E-0D1C-3986-8C95-48A020DD5F77 <155 126 119 118 108 88 6 5 4 3 1>

 

pmset -g assertions
2018-10-05 23:24:14 +0200
Assertion status system-wide:
   BackgroundTask                 0
   ApplePushServiceTask           0
   UserIsActive                   1
   PreventUserIdleDisplaySleep    0
   PreventSystemSleep             0
   ExternalMedia                  1
   PreventUserIdleSystemSleep     0
   NetworkClientActive            0

Listed by owning process:
   pid 57(powerd): [0x0000000600088002] 02:14:00 ExternalMedia named: "com.apple.powermanagement.externalmediamounted"
   pid 98(hidd): [0x00001ca400098506] 00:00:00 UserIsActive named: "com.apple.iohideventsystem.queue.tickle.4294968346.3"
    Timeout will fire in 600 secs Action=TimeoutActionRelease
Kernel Assertions: 0xc=USB,BT-HID
   id=500  level=255 0x4=USB mod=01/01/1970, 01:00 description=com.apple.usb.externaldevice.00600000 owner=IOUSBHostDevice
   id=502  level=255 0x4=USB mod=01/01/1970, 01:00 description=com.apple.usb.externaldevice.00800000 owner=DataTraveler 3.0
   id=503  level=255 0x4=USB mod=01/01/1970, 01:00 description=com.apple.usb.externaldevice.1d100000 owner=IOUSBHostDevice
   id=504  level=255 0x4=USB mod=01/01/1970, 01:00 description=com.apple.usb.externaldevice.1a100000 owner=IOUSBHostDevice
   id=507  level=255 0x4=USB mod=01/01/1970, 01:00 description=com.apple.usb.externaldevice.1d120000 owner=USB 2.0 Hub
   id=509  level=255 0x4=USB mod=01/01/1970, 01:00 description=com.apple.usb.externaldevice.1d121000 owner=Microsoft® 2.4GHz Transceiver v7.0
   id=510  level=255 0x4=USB mod=01/01/1970, 01:00 description=com.apple.usb.externaldevice.1d124000 owner=USB Receiv
   id=511  level=255 0x4=USB mod=01/01/1970, 01:00 description=com.apple.usb.externaldevice.1d123000 owner=BCM20702
   id=512  level=255 0x4=USB mod=01/01/1970, 01:00 description=com.apple.usb.externaldevice.00200000 owner=IOUSBHostDevice
   id=514  level=255 0x4=USB mod=01/01/1970, 01:00 description=com.apple.usb.externaldevice.00220000 owner=Microsoft® LifeCam Cinema(TM)
   id=515  level=255 0x8=BT-HID mod=01/01/1970, 01:00 description=com.apple.driver.IOBluetoothHIDDriver owner=BNBTrackpadDevice
Idle sleep preventers: IODisplayWrangler

 

pmset -g
System-wide power settings:
Currently in use:
 standby              0
 Sleep On Power Button 1
 womp                 0
 halfdim              1
 hibernatefile        /var/vm/sleepimage
 powernap             0
 autorestart          0
 networkoversleep     0
 disksleep            0
 standbydelayhigh     86400
 sleep                0
 autopoweroffdelay    28800
 hibernatemode        0
 autopoweroff         0
 ttyskeepawake        1
 displaysleep         10
 Standby Battery Threshold 50
 standbydelaylow      0

 

system_profiler SPSerialATADataType|grep TRIM
          TRIM Support: No

 

syslog.rtf

EFI.zip

Cezar’s Mac Pro ioreg.zip

AML from patchmatic.zip

Edited December 16, 2018 by iCezar

[Allan 5,177]

Try us this command:

sudo trimforce enable

 

[iCezar 1]

7 hours ago, Allan said:

Try us this command:

sudo trimforce enable

 

Thank you for the quick support. 

I actually had it enabled and disabled it to see if it helps. It didn't. I can enable it again if you believe it will have an impact on troubleshooting results

Edited October 6, 2018 by iCezar

[Allan 5,177]

TRIM is very important to the SSDs, using it you'll increase the life spam of your device.

[iCezar 1]

5 hours ago, Allan said:

TRIM is very important to the SSDs, using it you'll increase the life spam of your device.

Understood. But even with TRIM, the hangs and freezes were still there.

[Asher- 0]

I have just identified TRIM and ALPM as two definite stability culprits under High Sierra. With either enabled, exporting photos from DxO Photolab with 16GB RAM causes a crash before the first file completes. With 128GB RAM, crashes happen significantly later but still regularly.

 

With 16GB RAM and both disabled, I was able to export 1750 22.3mp DNGs before encountering a crash, which was possibly due to a corrupt file (not yet determined, but the file it was processing when it crashed was definitely corrupt). 

[Pavo 998]

42 minutes ago, Asher- said:

I have just identified TRIM and ALPM as two definite stability culprits under High Sierra. With either enabled, exporting photos from DxO Photolab with 16GB RAM causes a crash before the first file completes. With 128GB RAM, crashes happen significantly later but still regularly.

 

With 16GB RAM and both disabled, I was able to export 1750 22.3mp DNGs before encountering a crash, which was possibly due to a corrupt file (not yet determined, but the file it was processing when it crashed was definitely corrupt). 

If you are using APFS you should not be enabling TRIM, APFS has its own TRIM functions.

[ellaosx 184]

2 hours ago, Pavo said:

If you are using APFS you should not be enabling TRIM, APFS has its own TRIM functions.

@Pavo

Even it says NO in system profiler?

[iCezar 1]

Well, just enabled it back and the OS still hangs. It's also virtually impossible to start any 3rd party app , even if it was installed from the App Store. Also, I am having issues writing this message, the cursor hangs each 5 seconds for about 5 seconds

[inequals 1]

23 hours ago, Pavo said:

If you are using APFS you should not be enabling TRIM, APFS has its own TRIM functions.

First time I've seen this - I somehow ended up with APFS on my system drive (a NVME 960 Evo), are you telling me I should disable TRIM? For real? Or - should I just erase and restore my backup to HFS+, and keep TRIM enabled? Thanks in advance. 

[Pavo 998]

16 minutes ago, inequals said:

First time I've seen this - I somehow ended up with APFS on my system drive (a NVME 960 Evo), are you telling me I should disable TRIM? For real? Or - should I just erase and restore my backup to HFS+, and keep TRIM enabled? Thanks in advance. 

I can not tell you what file system you should be using, I can only tell you the facts between the 2. APFS has it's own TRIM function during the boot up phase (look at your kernel logs if you wanna verify "spaceman_trim"). HFS+ will only enable TRIM be default for "Apple" SSD types. 

[inequals 1]

3 minutes ago, Pavo said:

I can not tell you what file system you should be using, I can only tell you the facts between the 2. APFS has it's own TRIM function during the boot up phase (look at your kernel logs if you wanna verify "spaceman_trim"). HFS+ will only enable TRIM be default for "Apple" SSD types. 

I'm sorry I put the question in such a way that I was asking you to decide which FS I should use. I know that's a complex question with no proper answer. But, are you saying that APFS' TRIM implementation is such that it conflicts with having TRIM enabled elsewhere? Surely, any file system is operating at one level removed from the physical block management of the SSD? So the idea that APFS could replicate the entire TRIM functionality completely is confusing my stupid brain. 

As my system SSD is not "Apple", I'm leaning towards moving back to HFS+ w/TRIM forced on. 

[Pavo 998]

10 minutes ago, inequals said:

I'm sorry I put the question in such a way that I was asking you to decide which FS I should use. I know that's a complex question with no proper answer. But, are you saying that APFS' TRIM implementation is such that it conflicts with having TRIM enabled elsewhere? Surely, any file system is operating at one level removed from the physical block management of the SSD? So the idea that APFS could replicate the entire TRIM functionality completely is confusing my stupid brain. 

As my system SSD is not "Apple", I'm leaning towards moving back to HFS+ w/TRIM forced on. 

APFS will use its own TRIM function during the boot phase for all SSDs no matter if you enable the OS itself for TRIM or if its a "Apple" or "Non-Apple SSD. Essentially if you enable TRIM in the OS ie.... enable trim force command when the OS shuts down or reboots, it will go through its own function of the TRIM process. Then during the boot phase APFS itself will also go through its own TRIM process, this is bad because every TRIM process degrades the lifespan of SSDs by some unknown degree. HFS+ on the other hand relies on the OS to preform the TRIM process, which is why it requires you to force enable it on "Non-Apple" SSDs. But HFS+ is an out-dated FS and is soon to be deprecated. So I can not make this decision for you, you have to decide what FS you are going to use and how you are going to use it.

[ntsmkfob 20]

If someone with a real mac using an apple branded SSD could check their System Report and let us know if Trim is enabled on the disk hosting the APFS volume, that would be useful.

[ellaosx 184]

37 minutes ago, ntsmkfob said:

If someone with a real mac using an apple branded SSD could check their System Report and let us know if Trim is enabled on the disk hosting the APFS volume, that would be useful.

It is on my mbp & mba

[ntsmkfob 20]

9 hours ago, ellaosx said:

It is on my mbp & mba

Thanks!

[iCezar 1]

Thanks everyone for the replies. 

 

I tried a clean install on a different disk and then restored the installation on HFS, issues persist.

Updated all efi drivers.

Also removed a second SSDT for an older CPU that I have forgotten about after moving from Sandy Bridge to Ivy Bridge 

Any other thought?

Edited October 30, 2018 by iCezar

[iCezar 1]

UPDATE 2: uploading EFI folder

 

A little update here. Im not sure but I think this is related to some sort of new security measures Apple introduced in Mojave.

The issue only seems to happen when a network connection exists. 

 

I filmed the behaviour with and without a network connection and the result is pretty remarkable.

 

The launch time of a third party app is aprox. 1 minute when the computer is connected to the internet. 

The launch time when the connection is disabled is, well, instant, as it should be on a NVMe WD Black drive. 

 

Videos below:

 

- connection enabled: https://youtu.be/OzRkciFslUg

 

- connection disabled: https://youtu.be/1fRzUD6mtBk

 

 

EFI.zip

Edited December 16, 2018 by iCezar

[iCezar 1]

Biiiig update. I fixed the issue. It appears that blocking ocsp.apple.com on your firewall can cause Mojave to retry for almost 1 minute the verification of an applications certification validity with their OCSP server.
Don’t ask why I had it blocked, I just did.. ))

More about OCSP:

https://en.m.wikipedia.org/wiki/Online_Certificate_Status_Protocol

[inequals 1]

bugger. Just ping'd OCSP without issue, but I must say I haven't noticed the issue so much. I have a frankly ridiculous amount of startup apps enabled, and given that it seems normal now (either that or I have merely acclimatised to it). Anyhow, great find. Hopefully this will help others. Are you on the latest, 10.14.2?

[iCezar 1]

Are you on the latest, 10.14.2?

Yes.

 

I noticed that not only app launches were delayed but also simple shell commands (mount) which makes me believe that server receives a “little” more info than just cert validation requests...

 

Sent from my iPhone using Tapatalk